1.完整项目描述和程序获取
>面包多安全交易平台:https://mbd.pub/o/bread/Z56Vm59s
>如果链接失效，可以直接打开本站店铺搜索相关店铺：
>如果链接失效，程序调试报错或者项目合作也可以加微信或者QQ联系。
2.部分仿真图预览
3.算法概述
在现代通信系统中，高效可靠的信号检测至关重要。正交频分复用（OFDM）技术因其能有效对抗多径衰落、提高频谱效率等优点，被广泛应用于众多通信标准，如 4G、5G 移动通信以及数字视频广播等领域。正交相移键控（QPSK）则是一种常用的数字调制方式，它在有限的带宽内实现了较高的数据传输速率。然而，在复杂的通信环境中，OFDM-QPSK 信号会受到噪声、干扰等因素的影响，传统的信号检测方法在性能上存在一定的局限性。深度神经网络（DNN）以其强大的特征学习和模式识别能力，为 OFDM-QPSK 信号检测提供了新的解决方案，能够有效提升检测性能，适应复杂多变的通信场景。
4.部分源码
% 在数据符号位置插入 QPSK 信号
Sym_qpsk = func_QPSKMod(dataSym);
Msg_qpsk = reshape(Sym_qpsk, Nsubs, Nsymb);
%导频插入
Xpilot = 1 - 1j;
Loc_pilot = loc_pilot : It_pilot : NFrame;
Num_pilot_frame = length(Loc_pilot) * Nsubs;
Loc_data = 1 : NFrame;
Loc_data(Loc_pilot(:)) = [];
dat_ifft = zeros(NFFT-1,NFrame);
dat_ifft(:, Loc_pilot(:)) = Xpilot;
dat_ifft(:, Loc_data(:)) = Msg_qpsk;
dat_ifft = [zeros(1, NFrame); dat_ifft];
%OFDM发射
Tx_ofdm = func_TOFDM(dat_ifft, NFFT, NCP);
%Channel
OFDM_ch = (1 / sqrt(2)) * randn(Nsymb2, 1) + (1 / sqrt(2)) * 1j * randn(Nsymb2, 1);
OFDM_ch_sp = repmat(OFDM_ch, NFrame, 1);
Rx_ofdm0 = OFDM_ch_sp .* Tx_ofdm;
Rx_ofdm0 = awgn(Rx_ofdm0,SNR,'measured');
Hch_ideal = Rx_ofdm0 ./ OFDM_ch_sp;
...................................................................
% Deep learning
[Rx_ofdm_feature, ~] = func_ofdm_feature(Rx_ofdm, Msg_qpsk(1:8));
Rx_de_H_dnn = predict(DNN_Trained, Rx_ofdm_feature);
Rx_de_H_dnn2 = Rx_de_H_dnn(1:2:end, :) + 1j * Rx_de_H_dnn(2:2:end, :);
Rx3_de_qpsk = func_QPSKDemod(Rx_de_H_dnn2);
Rx3_de_qpsk2 = de2bi(Rx3_de_qpsk, 2);
Rx_de_qpsk_dnn = reshape(Rx3_de_qpsk2, [], 1);
.................................................
figure;
semilogy(SNRs,ber_ls,'-k<',...
'LineWidth',1,...
'MarkerSize',6,...
'MarkerEdgeColor','k',...
'MarkerFaceColor',[0.3,0.9,0.3]);
hold on
semilogy(SNRs,ber_mmse,'-bs',...
'LineWidth',1,...
'MarkerSize',6,...
'MarkerEdgeColor','k',...
'MarkerFaceColor',[0.9,0.0,0.0]);
hold on
semilogy(SNRs,ber_dnn,'-r>',...
'LineWidth',1,...
'MarkerSize',6,...
'MarkerEdgeColor','k',...
'MarkerFaceColor',[0.9,0.9,0.0]);
grid on
xlabel('SNR');
ylabel('BER');
legend('LS','MMSE','DNN');
figure;
semilogy(SNRs,ser_ls,'-k<',...
'LineWidth',1,...
'MarkerSize',6,...
'MarkerEdgeColor','k',...
'MarkerFaceColor',[0.3,0.9,0.3]);
hold on
semilogy(SNRs,ser_mmse,'-bs',...
'LineWidth',1,...
'MarkerSize',6,...
'MarkerEdgeColor','k',...
'MarkerFaceColor',[0.9,0.0,0.0]);
hold on
semilogy(SNRs,ser_dnn,'-r>',...
'LineWidth',1,...
'MarkerSize',6,...
'MarkerEdgeColor','k',...
'MarkerFaceColor',[0.9,0.9,0.0]);
grid on
xlabel('SNR');
ylabel('SER');
legend('LS','MMSE','DNN');
0X_075m
---
